Mitochondria-Associated Endoplasmic Reticulum Membrane Biomarkers in Coronary Heart Disease and Atherosclerosis: A Transcriptomic and Mendelian Randomization Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper was written according to the instructions for authors. The introduction represents an adequate overview of up to dateliterature. The methodology is clear and reproducible, and the results are well presented and correlated with similar studies.

1. There is no space between the period at the end of a sentence and the beginning of a new one. LINES: 16, 18, 28, 36. 2. Everywhere in the text (introduction and discussion) there is no space before the parentheses in which the references are written. 3. No space between period and parenthesis: LINE 253, 256, 259, 280, 281, 401. 4. The methodology lacks information on how the animals were euthanized (by anesthetic, exsanguination or some other method). Also, it wouldn't be a bad idea to briefly add what exactly a high-fat diet means and what a normal chow diet means. 5. In the methodology, state at which magnification the changes were observed and recorded. 6. In Figure 8 F-G, the scales on the microphotographs are missing, as well as the information at which magnification the picture was taken. Also, photomicrographs are very small. The suggestion is to make two separate plates for these two stainings, in order to more easily see the differences between the groups.

Author Response

Comments 1: There is no space between the period at the end of a sentence and the beginning of a new one. LINES: 16, 18, 28, 36.

Response 1:Thank you for your valuable suggestion. We have made the corresponding revisions in LINES 16, 18, 28, and 36 as you suggested.

Comments 2: Everywhere in the text (introduction and discussion) there is no space before the parentheses in which the references are written.

Response 2: Thank you for your suggestion. We have carefully checked the entire manuscript for this issue and made the necessary revisions accordingly.

Comments 3: No space between period and parenthesis: LINE 253, 256, 259, 280, 281, 401.

Response 3: Thank you for your suggestion. We have added the necessary spaces in the corresponding sections of the manuscript.

Comments 4: The methodology lacks information on how the animals were euthanized (by anesthetic, exsanguination or some other method). Also, it wouldn't be a bad idea to briefly add what exactly a high-fat diet means and what a normal chow diet means.

Response 4: Thank you for your valuable suggestions. We have revised the section "2.9. Development of animal models for atherosclerosis" according to your recommendations. The dietary composition and euthanasia method have been added as requested, with the modifications highlighted in red for your review. The revised content is as follows: The mice were randomly assigned into two groups (n=5 per group): the high-fat diet (HFD) group received a high-fat diet (D12108C, Research Diets, USA; containing 40% kcal from fat, 43% from carbohydrates, and 17% from protein) for 12 weeks, while the normal chow diet (NCD) group was fed standard chow (containing approximately 10% kcal from fat, 70% from carbohydrates, and 20% from protein) for 12 weeks. At the conclusion of the feeding period, mice were euthanized by CO₂ inhalation followed by cervical dislocation and arterial tissues were collected for subsequent analysis.

Comments 5: In the methodology, state at which magnification the changes were observed and recorded.

Comments 6:In Figure 8 F-G, the scales on the microphotographs are missing, as well as the information at which magnification the picture was taken. Also, photomicrographs are very small. The suggestion is to make two separate plates for these two stainings, in order to more easily see the differences between the groups.

Response 5 and 6: Thank you very much for your suggestion. We have added a scale bar to the figure to clearly indicate the size. The revised figure can be found in the updated manuscript submitted with this revision. Meanwhile, we have added the following content to the Methods section of the manuscript: Representative immunofluorescence images were captured at 200× magnification using a fluorescence microscope, with scale bars indicating 100 μm. The revised content is highlighted in red for your review. Thank you. 

Reviewer 2 Report

Comments and Suggestions for Authors

Overall, the manuscript addresses an important topic; however, several major issues need to be addressed to improve clarity, rigor, and scientific impact.
Major Comments:

• All of your figures are blurred and need to be uploaded in high quality. I had difficulty reading the content of the figures. The figures need to be replaced.

• Your Introduction does not adequately introduce the topic of your research. The Introduction cannot be summarized in only two paragraphs and needs to be expanded.

• Please explain in detail why the Support Vector Machine (SVM) was used. What was its purpose? Also, explain the construction of the SVM-based diagnostic model.

• I do not think Figure 7 adds much to the overall story. Docking-derived binding energies are predictive rather than definitive and should ideally be supported by experimental evidence such as ITC, SPR, or enzymatic assays. Unless additional evidence is provided, this figure can be removed to keep the manuscript focused.

• The data shown in Figures 8A, 8B, and 8C are one of the strengths supporting your biomarker findings; however, they were not validated properly. Showing expression of these biomarkers alone is insufficient. You must also show RNA and/or proteins involved in the maintenance of MAM. MFN1/2 appear to be the most relevant in the pathogenesis of CVD. Other proteins of interest include OPA1, which preserves the ER–mitochondria interface and is involved in ischemia/reperfusion (I/R), whereas inhibition of FIS1 and DRP1 has been reported to be cardioprotective.

• The study focuses on CHD, yet experimental validation employs an atherosclerosis mouse model. While relevant, CHD encompasses broader pathological mechanisms. The authors should clarify the rationale for this model choice and its relevance to the specific CHD phenotypes studied.
  
  Minor Comments:

• Please improve the English language and grammar throughout the manuscript.

• When explaining your findings, begin with hypothesis-driven sentences.

• Discuss the limitations of your study, such as the small sample size of human PBMCs.

• In the Discussion, avoid subheadings and instead link your findings into a single cohesive narrative.

• Discuss the mechanisms of mitochondrial regulation in cardiac physiology and cite this relevant literature, PMID: 41301698; PMID: 40164849; PMID: 38713090; PMID: 38474356.

• Explain how MAM functions, including discussion of the IP3R–GRP75–VDAC complex. PMID: 3562409.

• Cite more recent and impactful references.

Author Response

Comments 1:  All of your figures are blurred and need to be uploaded in high quality. I had difficulty reading the content of the figures. The figures need to be replaced.

Response 1: Thank you very much for your suggestion. We have uploaded all figures in higher resolution. Please find them in the revised manuscript. Thank you.

Comments 2: Your Introduction does not adequately introduce the topic of your research. The Introduction cannot be summarized in only two paragraphs and needs to be expanded.

Response 2: We greatly appreciate your valuable suggestion. We have substantially expanded the Introduction section to provide more comprehensive background information. The modifications are shown below.

Comments 3: Please explain in detail why the Support Vector Machine (SVM) was used. What was its purpose? Also, explain the construction of the SVM-based diagnostic model.

Response 3: Thank you for this important question that allows us to clarify our methodology. We would like to emphasize that in our study, we employed SVM-RFE (Support Vector Machine-Recursive Feature Elimination) as a feature selection algorithm rather than constructing an SVM-based diagnostic model. The primary purpose of using SVM-RFE was to identify the most relevant biomarker genes from our candidate gene list through an iterative recursive elimination process that removes the least important features based on the weight vector of the SVM classifier. We specifically selected SVM-RFE because it provides robust feature ranking when handling high-dimensional genomic data where the number of features may exceed the number of samples, and it effectively complements our consensus-based feature selection strategy. By combining SVM-RFE with LASSO and Boruta algorithms, we implemented a stringent validation approach where only genes identified by all three independent methods were retained as biomarker candidates, thereby minimizing false positives and enhancing the reliability of our final biomarker identification. These selected genes were then subjected to further Mendelian randomization validation tests including Steiger's direction test, heterogeneity test, pleiotropy test, and leave-one-out test to ensure their causal relationship with CHD. We apologize if our original description caused any confusion and have revised the relevant section to provide clearer explanation of this feature selection methodology.

Comments 4:    I do not think Figure 7 adds much to the overall story. Docking-derived binding energies are predictive rather than definitive and should ideally be supported by experimental evidence such as ITC, SPR, or enzymatic assays. Unless additional evidence is provided, this figure can be removed to keep the manuscript focused.

Response 4: Thank you very much for your suggestion. We have removed the molecular docking content from Figure 7. and all corresponding descriptions from the manuscript, including the molecular docking-related content and section 2.8 (Prediction of Targeted Drugs and Molecular Docking) in the Methods section, to make the results more focused.

Comments 5: The data shown in Figures 8A, 8B, and 8C are one of the strengths supporting your biomarker findings; however, they were not validated properly. Showing expression of these biomarkers alone is insufficient. You must also show RNA and/or proteins involved in the maintenance of MAM. MFN1/2 appear to be the most relevant in the pathogenesis of CVD. Other proteins of interest include OPA1, which preserves the ER–mitochondria interface and is involved in ischemia/reperfusion (I/R), whereas inhibition of FIS1 and DRP1 has been reported to be cardioprotective.

Response 5: Thank you very much for your suggestion. Building upon Figure 8, we have added additional evidence demonstrating the involvement of MAMs in the progression of atherosclerosis. Using cDNA obtained from reverse transcription of our previous patient samples, we performed supplementary detection of several MAM-related markers including MFN1/2 and Fis1. The results have been incorporated into the revised Figure 8. Please find the updated figure enclosed.

Comments 6: The study focuses on CHD, yet experimental validation employs an atherosclerosis mouse model. While relevant, CHD encompasses broader pathological mechanisms. The authors should clarify the rationale for this model choice and its relevance to the specific CHD phenotypes studied.

Response 6: We sincerely appreciate your insightful comment regarding our model selection, which raises an important methodological consideration. We fully acknowledge that CHD encompasses a broader spectrum of pathological mechanisms beyond atherosclerosis alone; however, our choice of the LDLR-/- atherosclerosis mouse model is based on sound scientific rationale. Atherosclerosis represents the fundamental pathological substrate in approximately 90% of clinical CHD cases and directly correlates with the clinical manifestations observed in our patient cohorts, including both chronic coronary syndrome and unstable angina subtypes as shown in Table 2. Our human validation cohort comprised patients with confirmed CHD whose conditions are primarily driven by underlying atherosclerotic disease, and the LDLR-/- model effectively recapitulates the key pathophysiological features observed in these patients, including lipid accumulation, endothelial dysfunction, and inflammatory responses. Furthermore, the mitochondria-associated ER membrane dysfunction mechanisms we identified—particularly lipid metabolism dysregulation, calcium homeostasis disruption, and inflammatory signaling—are especially relevant to atherosclerotic plaque formation and progression, which directly contribute to CHD pathogenesis. The LDLR-/- mouse model is widely recognized as a gold-standard experimental system for studying CHD-related vascular pathology, as evidenced by extensive validation in high-impact cardiovascular journals. Nevertheless, we recognize that this model may not fully recapitulate all aspects of clinical CHD, particularly acute coronary syndromes involving plaque rupture and thrombosis, or non-atherosclerotic mechanisms such as coronary vasospasm or microvascular dysfunction. To address this limitation transparently, we have added the following clarification to our Discussion section: "While our experimental validation employed an LDLR-/- atherosclerosis mouse model, this approach was selected based on atherosclerosis being the predominant underlying pathology in CHD, accounting for the majority of cases in our clinical cohort. The LDLR-/- model effectively recapitulates key pathophysiological features relevant to CHD, including endothelial dysfunction, lipid accumulation, and inflammatory responses. However, we acknowledge that this model may not capture all CHD phenotypes, particularly acute thrombotic events or non-atherosclerotic mechanisms. Future studies incorporating additional models, such as myocardial infarction or ischemia-reperfusion models, would provide complementary insights into the broader spectrum of CHD pathology and further validate the roles of DHX36 and GPR68 across different disease contexts." We believe this clarification appropriately addresses your concern while maintaining scientific rigor, and we hope this explanation strengthens the manuscript.

Minor Comments

Comments 7:•    Please improve the English language and grammar throughout the manuscript.

Response 7: Thank you for your valuable suggestions. We have carefully reviewed and revised the grammar and language throughout the manuscript. Please find the revised version attached. Should you have any further comments or suggestions, please do not hesitate to let us know.

Comments 8:When explaining your findings, begin with hypothesis-driven sentences.

Response 8:Thank you very much for your insightful comments. In response to your suggestions, we have revised our conclusions to present the findings in a more cautious and nuanced manner. We have revised the abstract as follows: Conclusion: This study provides evidence supporting a mechanistic link between MAM dysfunction and CHD pathogenesis, identifying candidate biomarkers that have the potential to serve as diagnostic tools and therapeutic targets for CHD. While the validated biomarkers offer valuable insights into the molecular pathways underlying disease development, additional studies are needed to confirm their clinical relevance and therapeutic potential in larger, independent cohorts. We have also made corresponding revisions to the conclusion section of the manuscript: 5. Conclusion In conclusion, this study presents initial findings that suggest a possible causal relationship between MRGs and CHD through a comprehensive analytical approach. By leveraging transcriptomic sequencing data, Mendelian randomization analysis, and machine learning algorithms, we have identified two potential candidate biomarkers, DHX36 and GPR68, that warrant further investigation. Our exploratory investigation into the biological pathways, immune cell involvement, and molecular mechanisms associated with these biomarkers has begun to enhance our understanding of the underlying pathophysiology of CHD and may indicate potential avenues for therapeutic strategies and clinical diagnostics. Importantly, the specific mechanisms and molecular targets identified in this study should be considered preliminary and require rigorous validation in larger, independent cohorts and diverse populations to establish their roles in the development and progression of CHD. This research serves as a starting point for future studies to explore the intricate interplay between mitochondrial function and cardiovascular health, with the ultimate aspiration of improving patient outcomes through targeted interventions and a better understanding of the genetic basis of CHD.

Comments 9:Discuss the limitations of your study, such as the small sample size of human PBMCs.

Response 9: Thank you very much for your suggestion. We have added the following description to the limitations section: Secondly, our study is limited by the small sample size of human PBMCs and the lack of associated clinical data, which restricts the applicability of our findings to real-world scenarios and prevents us from gaining deeper insights into their clinical implications.

Comments 10: In the Discussion, avoid subheadings and instead link your findings into a single cohesive narrative.

Response 10: We sincerely appreciate your valuable suggestion regarding the Discussion section structure. We have given this matter careful consideration and would like to respectfully discuss this point further with you. While we fully understand the merit of creating a cohesive narrative, we found that presenting the discussion without subheadings made the overall structure somewhat challenging to follow and compromised the logical clarity of our arguments. To enhance readability and maintain a clear organizational framework, we have intentionally incorporated subheadings to guide readers through our key findings and their implications. However, we remain open to your expertise and perspective on this matter. If you still recommend removing the subheadings in favor of a unified narrative approach, we will be more than happy to revise the manuscript accordingly. We simply wanted to engage in this dialogue to ensure we are making the most appropriate structural choice for effectively communicating our research findings. Once again, we thank you for your thoughtful feedback and guidance throughout this review process.

Comments 11:•    Discuss the mechanisms of mitochondrial regulation in cardiac physiology and cite this relevant literature, PMID: 41301698; PMID: 40164849; PMID: 38713090; PMID: 38474356.

Response 11：Thank you very much for your valuable suggestion. We have incorporated the references you recommended into the revised manuscript. The specific citations are as follows: 9.    Patyal, P.; Azhar, G.; Verma, A.; Sharma, S.; Shrivastava, J.; Abdi, S.A.H.; Zhang, X.; Wei, J.Y. Mitochondrial Dynamics in Aging Heart. Biomedicines 2025, 13, 2603, doi:10.3390/biomedicines13112603. 10.    Patyal, P.; Azhar, G.; Zhang, X.; Verma, A.; Wei, J.Y. Cardiac-Specific Overexpression of Serum Response Factor Regulates Age-Associated Decline in Mitochondrial Function. Geroscience 2025, 47, 6565–6582, doi:10.1007/s11357-025-01629-2. 11.    Balderas, E.; Lee, S.H.J.; Rai, N.K.; Mollinedo, D.M.; Duron, H.E.; Chaudhuri, D. Mitochondrial Calcium Regulation of Cardiac Metabolism in Health and Disease. Physiology (Bethesda) 2024, 39, 0, doi:10.1152/physiol.00014.2024. 12.    Patyal, P.; Zhang, X.; Verma, A.; Azhar, G.; Wei, J.Y. Inhibitors of Rho/MRTF/SRF Transcription Pathway Regulate Mitochondrial Function. Cells 2024, 13, 392, doi:10.3390/cells13050392.

Comments 12:•    Explain how MAM functions, including discussion of the IP3R–GRP75–VDAC complex. PMID: 3562409.

Response 12: Thank you very much for your suggestion. Regarding the reference with ID 3562409 that you mentioned, we suspect there may be a slight discrepancy, as we were unable to locate the corresponding literature using this identifier. However, as you recommended earlier, we have incorporated additional relevant references to strengthen this section, which are now included in the revised manuscript for your review. If there has been any modification to the PMID or if you could provide the correct identifier for this particular reference, we would greatly appreciate it if you could inform us at your earliest convenience. Thank you for your patience and guidance.

Comments 13:•    Cite more recent and impactful references.

Response 13: We sincerely appreciate your valuable suggestion. We have incorporated additional references as recommended.

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review this interesting manuscript. I would also like to thank the authors for their valuable work. I have these minor comments.

1- For the introduction, more information is needed. Readers would be interested in how MAMs work. A figure illustrating their function would increase impact and make it easier to understand the role of MAMs. 
2- The figures in Figure 1 are small and have low resolution. I highly suggest having high-resolution images for this.
3- For PCA, more information is needed.  PERMANOVA (R²) needs to be mentioned. And the P value. In the low-resolution image, it's clear there are two distinct groups, but these statistics are important to note. 
4- Almost all the figures in the manuscript are low resolution, and it's very difficult to read the font in many of them. Even when I zoom in, I still can't read the font in many of the figures. I highly suggest getting high-resolution ones.


Other than this, I've no other concerns regarding this study. 

Author Response

Comments 1: For the introduction, more information is needed. Readers would be interested in how MAMs work. A figure illustrating their function would increase impact and make it easier to understand the role of MAMs.

Response 1: Thank you very much for your suggestions. We have supplemented the introduction section with a schematic diagram illustrating the mechanisms and functions of MAMs. Please find the added figure below.

Comments 2: The figures in Figure 1 are small and have low resolution. I highly suggest having high-resolution images for this.

Response 2: Thank you very much for your suggestions. We have re-uploaded all figures with higher resolution. Please review them. Thank you.

Comments 3: For PCA, more information is needed. PERMANOVA (R²) needs to be mentioned. And the P value. In the low-resolution image, it's clear there are two distinct groups, but these statistics are important to note.

Response 3: Thank you for your valuable suggestion. The R² value in the PCA analysis was 0.12856 with a p-value of 0.001. We have now added these statistical parameters to the figure, which has been updated accordingly in the revised manuscript. The revised figure is presented below.

Comments 4: Almost all the figures in the manuscript are low resolution, and it's very difficult to read the font in many of them. Even when I zoom in, I still can't read the font in many of the figures. I highly suggest getting high-resolution ones.

Response 4: Thank you very much for your valuable suggestions. We have re-uploaded all figures with improved resolution. Please find them for your review. Thank you.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have sufficiently addressed my concerns, and I recommend this manuscript for acceptance.